test(console): name the focused element when the mobile drawer check fails

[freeqaz]

Problem

Finding 2016: Console Browser went red once on #2255 (run 39250245319, job 118052053777) at tests/browser/console.test.mjs:846, the unlabelled check that the reopened mobile drawer's Agents link holds focus right after the toggle click. It failed as false !== true, so nothing says what held focus instead. The test failed once in 1,734 runs, during the 10-19 runner slowdown.

Change

Test-only. A focusSettles(locator, settleMs) helper samples document.activeElement and describes it (tag, role, class, accessible text, detached or not).

• Opened-drawer check (the flake): waits up to 1 s for focus to reach the Agents link (or the Settings item in the account-menu case). If it settles after starting elsewhere, the test logs a diagnostic with the first sample. On failure the message names the target, what holds focus, the first sample and the drawer's aria-expanded.
• Post-release check: keeps its single strict sample (no settling) and now names what holds focus too.

openDrawer focuses the first nav link synchronously, so the 1 s wait only covers a click that lands while the shell is being replaced. A real focus move would still fail, and the message now says where focus went.

Verification

• Mutant: openDrawer with no focus call fails with The opened drawer focuses its navigation link <a> "◇Agents"; focus is on <body> (first check: <button class="mobile-toggle"> "Open navigation"; drawer true).
• Repro attempts: the three mobile-drawer tests 25/25 on 1 CPU and 20/20 on 1 CPU shared with a busy loop. No failure, and focus never settled late.
• Pinned at 2 CPUs (taskset): the three mobile-drawer tests 20/20. Full console.test.mjs 48/48. Repeated on the final head after the review nits: the class is read with getAttribute (an SVG element's className is an object), and the drawer state is read only on failure.

Risks

Low. Test-only. The opened-drawer check now tolerates up to 1 s of focus elsewhere before failing, and logs it when that happens. The post-release "keeps focus" check is unchanged.

[clawsweeper]

🦞👀
ClawSweeper picked this up.

Pull request received. I will update this pull request when review starts.

ClawSweeper review complete

ClawSweeper finished reviewing this revision. The review result is being finalized.

View the workflow run.

[clawsweeper]

Codex review: needs maintainer review before merge.

What this changes

This PR adds focused-element descriptions and bounded initial focus polling to two existing mobile-drawer browser tests.

Example: The mobile drawer opens but focus remains on the page body.

• Before: The assertion reports false !== true.
• After: The failure names the expected Agents link, the body holding focus, the first observed element, and the drawer state.

Review scores

Measure	Result	What it means
Overall readiness	🦐 gold shrimp (3/6)	A useful, focused test improvement with no identified defect; confidence rests on source review and reported browser-harness validation, with exact-head browser CI still running.
Proof confidence	🦐 gold shrimp (3/6)	This maintainer-authored test-only change is outside the ordinary live-proof gate; the body reports mutation sensitivity and browser-suite runs, while inspected screenshots concern the linked Secret-access feature and do not demonstrate drawer diagnostics.
Patch quality	🐚 platinum hermit (4/6)	No actionable review findings were identified.

Product

Kind: Test only · Worth it: Yes · Fix scope: Complete
User problem: Contributors investigating a mobile-drawer test failure cannot tell which element held focus.
Reason: Actionable failure diagnostics have clear maintenance value, and the maintainer-authored PR records the assertion-timing choice.

Merge readiness

✅ Ready for maintainer review

Keep this PR open: it provides useful browser-test diagnostics, preserves the strict post-render focus check, and introduces no identified defect.

Priority: P3
Reviewed head: 615d6b1bd1064a2e93bc0c427cee420ef141a50a

Before merge

None.

Findings

None.

Tests

• Missing end-to-end proof: The reported intermittent failure was not reproduced; this test-only change adds no production behavior requiring a new end-to-end scenario.

Agent review details

How this fits together

The browser suite opens the controller-backed Console with fixture storage and Drivers, holds a Namespaces response, and checks drawer state and keyboard focus around the resulting render.

flowchart LR
  A[Browser test] --> B[Controller fixture]
  B --> C[Console mobile drawer]
  C --> D[Focus samples]
  D --> E[Assertion diagnostics]

Loading

Technical review

Best possible solution:

Retain the focused diagnostics and bounded initial wait while preserving strict focus verification after the held response completes.

Do we have a high-confidence way to reproduce the issue?

The original intermittent focus failure has no reliable reproduction: the author reports unsuccessful constrained-CPU attempts, and the available job metadata establishes only that the cited browser job failed.

Is this the best way to solve the issue?

Extending the existing behavioral checks is proportionate, and the drawer contributor explicitly accepts bounded initial settling while preserving the consequential post-render assertion.

AGENTS.md: found and applied where relevant.

Codex review notes: model internal, reasoning medium; reviewed against 40676a7fed34.

Provenance checked

• Mobile-drawer focus assertions changes intended behavior with a stated reason (fix(console): keep an open mobile navigation drawer across a page load's second render #2300: Verify that a drawer reopened during loading retains its open state and focused control when fresh content renders.)

Testing

Proof path: in-process harness.

Security

None.

Evidence

What I checked:

• Verified introduced change: The pinned diff modifies only existing browser tests; initial focus gets bounded polling and the post-release check uses zero settling time. (tests/browser/console.test.mjs:879, 615d6b1bd106)
• Current-main behavior: Current main retains the unlabelled initial assertion and the original strict post-render assertion; these diagnostics remain a distinct contribution. (tests/browser/console.test.mjs:846, 40676a7fed34)
• Feature history and accepted scope: fix(console): keep an open mobile navigation drawer across a page load's second render #2300 established these two focus-preservation scenarios. Its author also authored this PR and explicitly records the initial one-second tolerance while retaining strict verification after rendering. (tests/browser/console.test.mjs:815, 40676a7fed34)
• Exact-head checks: Static Checks passed on the reviewed head; Console Browser and CodeQL were still running. No failing check was attributed to this diff. (615d6b1bd106)
• Reported intermittent failure: GitHub confirms the cited Console Browser job failed on the head of fix(console): grant Agent access to accepted Configuration Secrets #2255; job metadata alone does not establish the precise focus failure or its cause. (3cc2804741c3)
• Policy and inspection boundaries: Read the complete root policy, implementation status, and testing skills. No nested test instructions, maintainer-notes directory, or pr-simplify skill were present. Some historical blobs were unavailable with lazy fetching disabled; current source, blame, commit metadata, and original discussion remained readable. The checkout stayed clean and no tests or additional reviewer ran. (AGENTS.md:1, 615d6b1bd106)

Review metrics

Metric	Value	Why it matters
Change scope	Production +0/-0; tests +47/-6	The growth supplies shared sampling and diagnostics for existing cases without adding production seams or duplicate test groups.

Labels

Label changes:

No label changes.

Label justifications:

• P3: This improves contributor-facing test diagnostics without changing production behavior.
• rating: 🦐 gold shrimp: Overall readiness is 🦐 gold shrimp; proof is 🦐 gold shrimp and patch quality is 🐚 platinum hermit.
• status: 👀 ready for maintainer look: ClawSweeper has no concrete contributor-facing blocker left for this PR.

Rating scale

6/6 🦀 challenger crab · 5/6 🦞 diamond lobster · 4/6 🐚 platinum hermit · 3/6 🦐 gold shrimp · 2/6 🦪 silver shellfish · 1/6 🧂 unranked krab. Overall follows the weaker of proof and patch quality; ✨ marks media proof (a screenshot, video, or linked artifact) that directly shows the changed behavior.

Workflow

ClawSweeper edits this one comment on every review. Comment @clawsweeper re-review for a fresh review only. Repair and merge execution lanes are retired.

History

Review history (1 earlier review cycle)

• reviewed 2026-10-19T13:36:48.195Z sha 4f46a23 :: needs maintainer review before merge. :: none

Reviewed October 19, 2026, 9:45 AM ET / 13:45 UTC (Revision 2).

[freeqaz]

@clawsweeper re-review

[clawsweeper]

🦞👀
Exact review queued.

Re-review progress:

• State: Complete
• Detail: The durable review result and its route handoff completed.
• Run: https://github.com/openclaw/clawsweeper/actions/runs/39263018267
• Updated: 2026-10-19T13:46:40.977Z